CN105278524B - A kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system - Google Patents
A kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system Download PDFInfo
- Publication number
- CN105278524B CN105278524B CN201510760928.4A CN201510760928A CN105278524B CN 105278524 B CN105278524 B CN 105278524B CN 201510760928 A CN201510760928 A CN 201510760928A CN 105278524 B CN105278524 B CN 105278524B
- Authority
- CN
- China
- Prior art keywords
- model
- diagnostic
- information
- fault diagnosis
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 141
- 238000013459 approach Methods 0.000 title claims abstract description 14
- 238000012423 maintenance Methods 0.000 claims abstract description 48
- 238000012549 training Methods 0.000 claims description 100
- 238000000034 method Methods 0.000 claims description 11
- 238000004321 preservation Methods 0.000 claims description 10
- 238000012545 processing Methods 0.000 claims description 10
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 7
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 6
- 235000013399 edible fruits Nutrition 0.000 claims description 6
- 230000007774 longterm Effects 0.000 claims description 5
- 230000004048 modification Effects 0.000 claims description 4
- 238000012986 modification Methods 0.000 claims description 4
- 238000009434 installation Methods 0.000 claims description 3
- 230000001932 seasonal effect Effects 0.000 claims description 3
- 230000006870 function Effects 0.000 description 53
- 241001269238 Data Species 0.000 description 5
- 230000005856 abnormality Effects 0.000 description 4
- 230000007257 malfunction Effects 0.000 description 4
- 238000010606 normalization Methods 0.000 description 4
- 230000008439 repair process Effects 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000007246 mechanism Effects 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000013499 data model Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 244000131316 Panax pseudoginseng Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B23/00—Testing or monitoring of control systems or parts thereof
- G05B23/02—Electric testing or monitoring
- G05B23/0205—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults
- G05B23/0208—Electric testing or monitoring by means of a monitoring system capable of detecting and responding to faults characterized by the configuration of the monitoring system
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Testing And Monitoring For Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
In order to solve the problems existing in the prior art, the present invention provides a kind of opening Approach for Hydroelectric Generating Unit Fault Diagnosis systems, including：Unit information maintenance module, diagnostic model maintenance module, diagnostic task maintenance module, idagnostic logout enquiry module, artificial fault diagnosis module.Wherein：Unit information maintenance module includes unit information service portion and set state service portion；Diagnostic model maintenance module includes that diagnostic model is safeguarded, model parameter is arranged, model sample is trained, model result uses part；Diagnostic task maintenance module includes that timed task is safeguarded, diagnoses unit maintenance, diagnostic log query portion；Idagnostic logout enquiry module includes idagnostic logout query portion and prediction record queries module；Artificial fault diagnosis module includes artificial fault diagnosis and Artificial Diagnosis query portion.The present invention can carry out fault diagnosis and prediction towards different machine set types, operating mode, fault diagnosis algorithm, sample data, can improve the safety and reliability of Hydropower Unit.
Description
Technical field
The invention belongs to equipment running status monitoring technical fields, are related to a kind of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system
System.
Background technology
For water power as clean reproducible energy, traffic control is flexible, the energy that each state all develops water power as priority.Closely
The China Nian Lai hydropower installed capacity rapid development, power station main equipment sharply increase, and maintenance workload increases severely, and service personnel is in short supply to be asked
Topic becomes increasingly conspicuous.Hydropower Unit failure and accident occur frequently simultaneously, such as rotor rupture, thrust bearing oil leak, guide bearing tile kilning, move not
Balance excessive, water logging workshop etc..
In order to understand Hydropower Unit operation conditions in time, catastrophic failure is avoided, major part power station is all installed at present
On-Line Monitor Device, can to the floor datas such as the vibration of Hydropower Unit, throw, temperature, voltage, electric current, active and reactive into
Row acquisition in real time, and alarm the quantity of state to cross the border.This mode has effectively reacted the present situation of unit, and can be right
Out-of-range conditions amount is alarmed, but is a lack of the early prediction to failure and identification, is unfavorable for the preventive maintenance of Hydropower Unit
And maintenance.
Invention content
In order to solve problem above, the present invention provides a kind of opening Approach for Hydroelectric Generating Unit Fault Diagnosis systems, including：Unit
Maintenance of information module, diagnostic model maintenance module, diagnostic task maintenance module, idagnostic logout enquiry module, artificial fault diagnosis
Module.Wherein：
The unit information maintenance module includes unit information service portion and set state service portion.
Further, the unit information service portion is to realize the typing of Hydropower Unit information by interface, change, delete
It is operated except equal.
Preferably, interface refers to the function interface of system.
Preferably, CS or BS structures may be used in interface system, can be realized by other programming languages such as Java or C#
Function interface.
Further, unit information includes power station title, machine group #, group name, rated power, unit model, throws
Produce the essential informations such as date, manufacturing firm, installation unit.
Further, the set state service portion is to realize the typing of set state information by interface, change, delete
It is operated except equal.
Preferably, whether set state information includes state encoding, Status Name, handles, suggests taking measures, shape occurs
State reason.
Further, state encoding is the target train value in fault sample, is numeric type data.
Further, if processing, is Boolean type data, and whether expression is remembered the result in fault diagnosis or prediction
Record is got off.If recorded, it can provide together and suggest taking measures and generating state cause information.
Further, what set state recorded in safeguarding is the various states of unit, including the normal condition of unit and each
The definition of kind abnormality, malfunction.
Preferably, to various abnormalities, malfunction, if processing should be set as true, and fill in and the state occurs
The reason of and the suggestion and measure that should take, to instruct the processing to failure.
The diagnostic model maintenance module includes that diagnostic model is safeguarded, model parameter is arranged, model sample is trained, model knot
Fruit such as uses at the parts.
Further, the diagnostic model service portion is by interface operation, typing fault diagnosis model information.
Preferably, fault diagnosis model information includes model name, model class name, training function name, anticipation function name, mould
Whether type description uses.Wherein model class name is external compiled class name or Service name；Anticipation function name and training letter
Several is external system or the compiled custom function name or interface name of service.
Preferably, the software translatings such as Matlab, Octave, R, Python training algorithm and prediction algorithm can be used.
Preferably, Service name can be the address of messenger service or Web Sevice services.
Further, the model parameter setting unit is that the parameter information of model is arranged by interface, model parameter letter
Breath includes the contents such as affiliated model, parameter name, parameter value, parameter type, parameter description, reference order.
Preferably, in calling model training or model prediction, the parameter list of affiliated model can be packaged into json
Character string, format are { parameter names 1:Parameter value 1, parameter name 2:Parameter value 2 ... }, each parameter is a name-value pair, more
It is separated with English comma between a parameter.
Further, model sample training part passes through setting training file and correlation model parameters, analyzing and diagnosing
The accuracy rate of model realizes the selection and calibration to diagnostic model.
Further, diagnostic model is the model information of typing in the above process, and training file is for selecting local csv
File, this document include each operational parameter data under some operating mode of Hydropower Unit, and such as vibration, voltage, electric current, is born temperature
The various data such as lotus and dbjective state.Csv is simple text file, is made of ranks, and often capable end is accorded with new line
It separates, is separated by comma between each column, excel or other tool open can be used.
Preferably, can require upload csv files the first row must be Hydropower Unit each operating parameter title and
Set state, and set state must necessarily be placed in first row or last row, all other row from the second row must all be several
Value type data.
Preferably, system all may be used using open data model as long as thinking operating parameter influential on set state
To be input to model data, set state value, that is, set state encoded radio is defined in set state information.Unit shape
Whether state information includes state encoding, Status Name, handles, suggests taking measures, generating state reason.Set state generally wraps
The institute included under unit operating mode is stateful, that is, includes the states such as the various failures, defect, alarm of unit, also includes the normal of unit
State.
Preferably, prediction ratio can be cured as 0.5,0.6,0.7,0.8,0.9,1, present in Hydropower Unit, that is, system
Each Hydropower Unit title, unit operating mode are generally divided into power generation, draw water, the phase modulation that draws water, turbine condenser mode etc., and normalization maximum value can be with
1,10,50,100,500,1000 are cured as, default value 100.
Further, the external trainer function where model training meeting calling model, and relevant parameter is passed to function ginseng
In number.
Preferably, automatic training mode can be used in model training：The operation that system can include according to training file first is joined
Number selects corresponding training pattern and relevant parameter is inputted respectively in the function of each training pattern；Each training pattern is finished
After export training result, including：Accuracy rate, sample size, predicted quantity；Then the training mould of 98% or more accuracy rate is preserved
Type, while by adjusting model parameter or being normalized or adjusting model sample training data and simulated again, it protects
Deposit the training pattern that accuracy rate is close or equal to 100%；When being less than 98% such as each training pattern accuracy rate, system passes through tune first
Mould preparation shape parameter is normalized or adjusts model sample training data and simulated again, preserve accuracy rate it is close or
Training pattern equal to 100%；If simulation accuracy rate is still less than 98% again, then system can select other training patterns to repeat
Above-mentioned simulation process preserves the training pattern that wherein accuracy rate is close or equal to 100%.
Preferably, active training pattern can be used in model training：User selects diagnostic model and training file, and is arranged pre-
Whether survey ratio prediction label, Hydropower Unit, set state, the related training parameter such as normalizes and is trained, and training pattern is held
Training result is exported after row, including：Accuracy rate, sample size, predicted quantity；Then the instruction of 98% or more accuracy rate is preserved
Practice model, while by adjusting model parameter or being normalized or adjust model sample training data progress mould again
It is quasi-, it is ensured that accuracy rate can be close or equal to 100% training pattern and preserve training result；If by adjusting model parameter,
The settings such as normalized, cannot allow model training rate of accuracy reached to 98% when, then prompt user replace another diagnosis mould
Type or the realization of adjustment model algorithm or adjusting training sample data, it is ensured that model training accuracy rate is close or equal to 100%.
Further, the model is using being believed by the model training result preserved during the model training of front
Breath, the fault diagnosis model for selecting a highest Hydropower Unit of accuracy rate to be used under specific operating mode, with the diagnostic model
Carry out fault diagnosis and prediction.
Preferably, model result information includes model name, model file, the model training as uploaded in model training
Whether file sample size, predicted quantity, accuracy rate, group name, unit operating mode, normalizes, normalizes maximum value, unit
Whether the target column of status Bar, as fault diagnosis and prediction prediction ratio, uses model.Other than whether using model, institute
There is information to be all from model training result and model information.Whether using model result value if it is true, illustrate the Hydropower Unit
Fault diagnosis is carried out using the model and prediction, the same Hydropower Unit, unit operating mode, training sample file are determined under the operating mode
Determine one and uses diagnostic model.
The diagnostic task maintenance module includes the parts such as timed task is safeguarded, diagnosis unit maintenance, diagnostic log are inquired.
Further, the timed task service portion includes timing failures diagnostic task and timing failures prediction task；
The diagnostic model that the timed task can be selected according to prefixed time interval timing operation front, and it is associated by diagnostic task
Unit is diagnosed, the current operating conditions parameter of the Hydropower Unit is obtained, fault diagnosis and forecast analysis are carried out to Hydropower Unit.
Further, timing task information include task names, task description, execute program name, execute the period, whether
Use the contents such as, operating status.
Preferably, the detailed process of timing failures diagnostic task can be：Available diagnostic task is first checked for, that is, is diagnosed
Program class name whether there is, and diagnostic task information inputs in timed task maintenance function, and the diagnosis for being stored in database is appointed
It is engaged in table, diagnostic task table records in direct searching database herein, if it does not exist, then prompting to build in log information
Vertical diagnostic task simultaneously exits Current Diagnostic task, and wherein log information is also stored in database table.It is examined if there is available
Disconnected mission bit stream, then continue checking for available diagnostic model and whether there is, diagnostic model information preservation is in database table, by examining
Disconnected model maintenance function typing.If there is no available diagnostic model information, then diagnosis mould is established in prompt in log information
Type simultaneously exits Current Diagnostic task.If there is available diagnostic model information, then whether available model result is continued checking for
It is the model information preserved after model training and training result in the presence of, model result.If available model result is not present,
Preservation model result is prompted in log information and exits Current Diagnostic task.If there is available model result information, then
It continues checking for Hydropower Unit state recording whether there is, if it does not exist, then prompting typing set state note in log information
It records and exits Current Diagnostic task.If there is Hydropower Unit state recording, then the available mould for the Hydropower Unit to be diagnosed is recycled
Type result records, model result record is read comprising model training file and model information according to model training file content
The first row trip information organizes Hydropower Unit "current" model operation data, and generates csv data files to be predicted, root
According to model relevant information and file to be predicted, calling model anticipation function carries out fault diagnosis, finally divides prediction result
Analysis and record, i.e., each prediction result compared with the state encoding in set state information, if prediction result is compiled with state
Code is consistent, illustrates that the prediction result of the Hydropower Unit is exactly the meaning that the state encoding represents, checks in the set state information
Whether whether processing is true, if it is true, need result information being recorded in idagnostic logout result, information includes diagnosis
Time, Status Name, suggestion takes measures, generating state reason does not record then if NO.In a prediction result, go out
Only retain first when existing identical prediction result, if prediction result does not appear in the meeting record log in set state information
Information, and prompt unit status information imperfect, ask typing to be encoded to the set state information of prediction result.
Preferably, the detailed process of timing failures prediction task can be：The prediction that is available for being first begin to check is appointed
Business, i.e. Prediction program class name whether there is, and then continue to check whether the information such as diagnostic model, model result, set state deposit
.Cycle will predict the model result of Hydropower Unit if all existing, per diem, week, three periods of the moon count nearest 60 periods
The average value of Hydropower Unit operational parameter data under identical operating mode.Then time series algorithm is used, including：Long-term trend,
Seasonal move, cyclical variations, erratic variation addition model, 60 day of prediction, week, after the moon 3 periods data.To prediction
As a result csv data files to be predicted are preserved into, according to model relevant information and file to be predicted, calling model anticipation function into
Row fault diagnosis is finally analyzed and is recorded to prediction result, and analysis result is stored in prediction result information, is remembered for prediction
Record query function inquiry.
Preferably, the definition of Cron expression formulas may be used in the period.
Preferably, the timing failures diagnostic task execution period is 5 minutes primary, and timing failures predict that the task execution period is
It executes 2 times daily.
Further, the diagnosis unit maintenance part is by interface operation, and maintenance will carry out fault diagnosis and prediction
Hydropower Unit set.
Preferably, diagnosis unit information include diagnostic task, whether diagnose, group name, unit model, rated power,
The contents such as date of putting into operation.Diagnostic task is the diagnostic task established during front diagnostic task is safeguarded, group name, unit model,
Rated power, date of putting into operation are unit essential informations present in system.
Further, the diagnostic log query portion be by interface operation, inquiry system execute fault diagnosis and
When failure predication task, the various log informations of record.
Preferably, log information content includes task names, logging level, logging time, log content etc., convenient for understanding
Diagnostic task executive condition.
Preferably, logging level can be divided into：Information, warning, mistake three grades.
The idagnostic logout enquiry module includes active inquiry and passive notification portion.
Further, active inquiry part is：User executes the diagnosis after fault diagnosis by interface operation inquiry system
As a result, diagnostic result content includes group name, set state, Diagnostic Time, diagnostic result.
Further, passive notification portion is：System discovery unit there are when failure, use by meeting automatic prompt during diagnosis
Family failure happening part, and the measure that prompts suggestion to take and the reason of may occur.
The artificial fault diagnosis module includes artificial fault diagnosis and Artificial Diagnosis query function.
Further, the artificial fault diagnosis part is active diagnosis of partial：User selects available diagnostic model simultaneously
Diagnostic file is uploaded, click starts to diagnose button to diagnostic file progress fault diagnosis.It is required that the fault diagnosis file uploaded must
It must be consistent with the training file format in the fault diagnosis model that uses, that is, require have identical columns, what each column represented
Meaning is consistent.The anticipation function of artificial fault diagnosis meeting calling model carries out fault diagnosis, and exports failure diagnosis information.
Current artificial failure diagnosis information can be shown in fault diagnosis result.Fault diagnosis result includes group name, unit shape
State, Diagnostic Time, diagnostic result, suggestion take measures, may occurrence cause.
Further, the Artificial Diagnosis inquiry is to inquire previous artificial fault diagnosis result by interface operation.Every time
Artificial fault diagnosis result can be saved, herein only according to group name, set state, Diagnostic Time section, diagnosis
As a result equal condition queries historical results record.Artificial Diagnosis inquiry content includes group name, set state, Diagnostic Time, examines
Disconnected result, suggest taking measures, may occurrence cause etc..
Diagnostic system of the present invention has the following advantages：
1. can be carried out towards different machine set types, different unit operating modes, different faults diagnosis algorithm, different sample datas
Fault diagnosis and prediction.It is real by providing open fault diagnosis model, fault sample training, periodic diagnosis task dispatching mechanism
Now accurate periodically fault diagnosis and forecast function, can effectively find failure in advance, and advance row specific aim occurs in failure
Repair can significantly improve the safety and reliability of Hydropower Unit, reduce maintenance cost.
2. general fault diagnosis model is established by the various operational parameter datas of Hydropower Unit using computer technology,
Realizing that Hydropower Unit is automatically short-term, --- 3 days following, medium-term and long-term --- the 3 weeks following, fault diagnosis in March and forecast function are
Power management personnel grasp the operating status of Hydropower Unit in advance, and timely, science formulates repair schedule, reduce maintenance cost, carry
High equipment safety is horizontal.
Description of the drawings
Fig. 1 systems function diagrams；
Fig. 2 timing failures diagnostic flow charts；
Fig. 3 timing failures predict flow chart.
Specific embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below in conjunction with attached drawing, to the present invention into
One step is described in detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to limit this
Invention.
Below in conjunction with attached drawing, the present invention is described in more detail.
In the present invention, the Hydropower Unit and set state information used in unit information maintenance module input system；
Diagnostic model maintenance module establishes open fault diagnosis model, and is trained by model sample, determines that each Hydropower Unit is each
Diagnostic model to be used under operating mode；Diagnostic task maintenance module realizes that the automatic diagnostic task of Hydropower Unit and automatic Prediction are appointed
The foundation and operation of business, and can check the daily record situation of task；Idagnostic logout inquires the automatic diagnosis for inquiring Hydropower Unit
The implementing result of task and prediction task understands Hydropower Unit operating status in time and in advance；Artificial fault diagnosis module is used for
The sample data of manual operation Hydropower Unit carries out fault diagnosis, and query history diagnostic result.
Fig. 1 is the systems function diagram of the present invention, above-mentioned technical purpose that the invention is realized by the following technical scheme：One kind is opened
Type Approach for Hydroelectric Generating Unit Fault Diagnosis system is put, includes mainly that unit information is safeguarded, diagnostic model is safeguarded, diagnostic task is safeguarded, diagnosis
Record queries, artificial fault diagnosis module.Wherein：
Unit information maintenance module includes that unit information is safeguarded and set state service portion.Unit information maintenance is to pass through
Interface realizes that operations, the interface systems such as typing, modification, the deletion of Hydropower Unit information can be CS or BS structures, can pass through
Other programming languages such as Java or C# are realized.Unit information includes power station title, machine group #, group name, rated power, machine
The essential informations such as group model, commissioning date, manufacturing firm, installation unit.Wherein, set state maintenance is to realize machine by interface
The operations such as typing, modification, the deletion of group status information.Whether set state information includes state encoding, Status Name, handle,
It is recommended that taking measures, generating state reason.State encoding in set state maintenance is the target train value in fault sample, is several
Value type data.Whether processing is Boolean type data, and whether expression is recorded the result in fault diagnosis or prediction.Such as
Fruit is recorded, and can provide together and suggest taking measures and generating state cause information.What set state recorded in safeguarding is machine
The various states of group, include the definition of the normal condition of unit and various abnormalities, malfunction.To various abnormalities,
Malfunction, if processing should be set as true, and fill in the reason of state occurs and the suggestion and measure that should be taken, to refer to
Lead the processing to failure.
Diagnostic model maintenance module includes that diagnostic model is safeguarded, model parameter is arranged, model sample is trained, model result makes
With equal parts.Wherein, diagnostic model maintenance is by interface operation, typing fault diagnosis model information, fault diagnosis model letter
Whether breath includes model name, model class name, training function name, anticipation function name, model description, uses.Model class name is outer
Portion it is compiled by the software translatings such as Matlab, Octave, R, Python it is good comprising training algorithm and prediction algorithm
Program class name, messenger service Service name, include the Service names of the addresss of service Web Sevice, anticipation function name and training function
Name is external system or the compiled custom function name or interface name of service.By taking Matalb grammers as an example, training mould
Type function defines the format that need to follow following parameter and return value：
Function retVal=funName (fileName, ycbl, ycbq, gyh, gyhmax, opt)
RetVal representative function return values, character string type, including information be training samples number, predicted quantity, standard
True rate, centre pass through English ":" separate；
FileName indicates to carry out the csv file names of model training, containing file path；
Ycbl indicates prediction ratio, belongs to the number between 0.5-1, and such as 0.5 indicates to predict a half data, and 1 is all predictions；
Ycbq indicates that prediction label, value 1 or -1 indicate first row if it is 1, and -1 indicates last row；
Gyh indicates whether to be normalized, value 1 or 0, and if it is 1, expression will be normalized, 0 table
Show without normalized.
Gyhmax indicates normalized maximum value, and value 1-1000, acquiescence is 100.
Opt indicates training parameter, string format, using json formats, such as { a:1,b:2 }, the value of expression parameter a is 1,
The value of b is 2.
The definition of prediction model function need to follow the format of following parameter and return value：
Function outLbl=funName (fileName, ycbq, gyh)
OutLbl representative function return values, character string type are the prediction results to input file fileName per a line,
Prediction result is numeric type data, and centre passes through English ":" separate；
FileName indicates to carry out the csv file names of model prediction, containing file path；
Ycbq indicates prediction label, is the value being arranged in model training, and value is 1 or -1；
Gyh indicates whether to be normalized, and is the value being arranged in model training, value 1 or 0.
Have to typing completely external class name or the Service name, while training function name and anticipation function at model class name
Name cannot be sky.The training function name and anticipation function name being actually typing only need to fill in funName, as LineTrain,
LinePredict.Program can according to conventional web services Naming conventions, such as comprising oblique stroke //, be external to distinguish model class name
Class or web services, can using the reflex mechanism of programming language according to model class name, function name, function parameter come organization procedure
Sentence call function or service so that model sample is trained and model prediction is carried out.
The model parameter setting is that the parameter information of model is arranged by interface, and model parameter information includes affiliated mould
The contents such as type, parameter name, parameter value, parameter type, parameter description, reference order.In calling model training or model prediction
When, the parameter list of affiliated model need to be packaged into json character strings, format is { parameter name 1:Parameter value 1, parameter name 2:
Parameter value 2 ... }, each parameter is a name-value pair, is separated with English comma between multiple parameters.
The model sample training is to select diagnostic model, training file, prediction ratio, pre- mark by interface operation
Whether label Hydropower Unit, unit operating mode, normalize, normalize the contents such as maximum value.Diagnostic model is the model letter of front typing
Breath, training file include each operating parameter number under some operating mode of Hydropower Unit for selecting local csv files, this document
According to, such as vibration, temperature, voltage, electric current, the various data of load and dbjective state.Csv is simple text file, by ranks
Composition, often capable end new line symbol separate, are separated by comma between each column, can use excel or other tool open.
It is required that the first row of the csv files uploaded must be each operating parameter title and set state of Hydropower Unit, and set state
First row or last row are must necessarily be placed in, all other row from the second row must all be numeric type data.System is used and is opened
The data model put may be input into model data, unit as long as thinking operating parameter influential on set state
State value, that is, set state encoded radio, is defined in set state information.Set state information includes state encoding, state name
Claim, whether handle, suggesting taking measures, generating state reason.Set state generally comprise under unit operating mode institute it is stateful, i.e.,
Including the states such as the various failures of unit, defect, alarm, also include the normal condition of unit.Prediction ratio is cured as 0.5,
0.6,0.7,0.8,0.9,1, each Hydropower Unit title present in Hydropower Unit, that is, system, unit operating mode be generally divided into power generation,
It draws water, the phase modulation that draws water, turbine condenser mode etc., normalization maximum value is cured as 1,10,50,100,500,1000, default value 100.
After each data item is selected in model sample training, you can train button, model training button that can call mould with click model
External trainer function where type, and relevant parameter is passed in function parameter.After external function executes training function, it can export
Training result, i.e. accuracy rate, sample size, predicted quantity, user can be confirmed whether to preserve this according to the height of accuracy rate
Model training result.General accuracy rate can click preservation model button, to model information and model training knot 98% or more
Fruit is preserved.If model accuracy rate is less than 95%, other training patterns or adjustment model parameter can be selected or returned
One changes processing or adjustment model sample training data, and by above method, the accuracy rate of model training result can reach
100%.Preservation model is required in external model training function as a result, also needing to preserve normalization if having selected normalization
Treated as a result, model result file name and normalized destination file title can be customized freely, but be required in model
Energy automatic identification comes out and calls in anticipation function.General both of these documents are stored in the catalogue at the same level of model sample data file
Under.
The model result determines Hydropower Unit in specific operating mode using being selected the training pattern after preservation
It is lower which diagnostic model to carry out fault diagnosis and prediction using.Model result information includes model name, model file, sample number
Whether amount predicted quantity, accuracy rate, group name, unit operating mode, normalizes, normalizes maximum value, set state row, prediction
Whether ratio uses model.Whether other than using model, all information are all from model training result and model information.Its
In, set state is classified as fault diagnosis and the target column of prediction；Whether using model result value if it is true, illustrate the hydroelectric machine
Group carries out fault diagnosis and prediction, the same Hydropower Unit, unit operating mode, training sample file under the operating mode using the model
Determine that uses a diagnostic model.
Diagnostic task module includes the parts such as timed task is safeguarded, diagnosis unit maintenance, diagnostic log are inquired.Wherein, fixed
When task maintenance be that timing mission bit stream is established by interface operation, by task execution period of setting and execute journey
Sequence allows system to execute program according to the execution periodic duty of setting automatically.Timing task information includes that task names, task are retouched
It states, executes program name, execute the period, whether use, the contents such as operating status.Period is defined using Cron expression formulas, Cron
Expression formula can indicate that extremely complex plan target executes the time.System should realize two basic timed task functions, one
A is timing failures diagnostic task, and a timing failures predict that task, the realization content of two tasks are detailed in flow chart below
It describes in detail bright.The timing failures diagnostic task execution period is 5 minutes primary, and timing failures predict that the task execution period is daily executes
2 times.Diagnosis unit maintenance is by interface operation, and maintenance will carry out fault diagnosis and the Hydropower Unit set of prediction.Diagnostic machine
The contents such as whether group information includes diagnostic task, diagnose, group name, unit model, rated power, date of putting into operation.Diagnosis is appointed
Business is the diagnostic task established during front diagnostic task is safeguarded, group name, unit model, rated power, date of putting into operation are to be
Unit essential information present in system.Diagnostic log inquiry is by interface operation, and inquiry system is executing fault diagnosis and event
When hindering prediction task, the various log informations of record.Log information content include task names, logging level (information, warning,
Mistake), logging time, log content etc., convenient for understanding diagnostic task executive condition.It includes that idagnostic logout is looked into diagnose enquiry module
Ask and predict record queries function.The idagnostic logout inquiry is by interface operation, and inquiry system is held in timing diagnostic task
When row fault diagnosis, the inquiry to diagnostic result.Diagnostic result content includes group name, set state, Diagnostic Time, diagnosis
As a result, suggest taking measures, may occurrence cause.The prediction record queries are by interface operation, and inquiry system is in timing
When predicting task execution prediction, the inquiry to prediction result.Prediction result content includes group name, set state, prediction day
Phase, prediction result, suggestion take measures, may occurrence cause.
It includes artificial fault diagnosis and Artificial Diagnosis query portion that artificial failure, which examines module,.Wherein, artificial fault diagnosis is
It by interface operation, selects available diagnostic model and uploads diagnostic file, click starts to diagnose button to diagnostic file progress
Fault diagnosis.It is required that the fault diagnosis file uploaded must be consistent with the training file format in the fault diagnosis model that uses
, that is, require there is identical columns, the meaning that each column represents to be consistent.The anticipation function of artificial fault diagnosis meeting calling model
Fault diagnosis is carried out, and exports failure diagnosis information.Current artificial failure diagnosis information can be shown in fault diagnosis result.
Fault diagnosis result includes group name, set state, Diagnostic Time, diagnostic result, suggestion takes measures, original occurs for possibility
Cause.Artificial Diagnosis inquiry is to inquire previous artificial fault diagnosis result by interface operation.Each artificial fault diagnosis result
It can be saved, herein only according to condition queries history such as group name, set state, Diagnostic Time section, diagnostic results
As a result it records.Artificial Diagnosis inquiry content includes that group name, set state, Diagnostic Time, diagnostic result, suggestion are taken and arranged
Apply, may occurrence cause etc..
Diagnostic system of the present invention has the following advantages：
1. can be carried out towards different machine set types, different unit operating modes, different faults diagnosis algorithm, different sample datas
Fault diagnosis and prediction.It is real by providing open fault diagnosis model, fault sample training, periodic diagnosis task dispatching mechanism
Now accurate periodically fault diagnosis and forecast function, can effectively find failure in advance, and advance row specific aim occurs in failure
Repair can significantly improve the safety and reliability of Hydropower Unit, reduce maintenance cost.
2. general fault diagnosis model is established by the various operational parameter datas of Hydropower Unit using computer technology,
Automatic short-term, the medium-term and long-term fault diagnosis and forecast function for realizing Hydropower Unit grasp hydroelectric machine in advance for power management personnel
The operating status of group, timely, science formulate repair schedule, reduce maintenance cost, and it is horizontal to improve equipment safety.
According to one embodiment of present invention, Fig. 2 is timing failures diagnostic flow chart, i.e., timing failures diagnostic task is interior
Portion's implementation flow chart.Timing diagnostic task is an executable program, is triggered and is executed by timed task.Internal implementation process is as follows
It is described.Available diagnostic task is first checked for, i.e. diagnostic program class name whether there is, and diagnostic task information is safeguarded in timed task
It inputs, and is stored in the diagnostic task table of database in function, diagnostic task table record is in direct searching database herein
Can, if it does not exist, then prompt establishes diagnostic task and exits Current Diagnostic task in log information.If there is available
Diagnostic task information then continues checking for available diagnostic model and whether there is, diagnostic model information preservation in database table, by
Diagnostic model maintenance function typing.If there is no available diagnostic model information, then diagnosis is established in prompt in log information
Model simultaneously exits Current Diagnostic task.If there is available diagnostic model information, then continuing checking for available model result is
No presence, model result are the model information preserved after model training and training result.If available model result is not present,
Preservation model result is then prompted in log information and exits Current Diagnostic task.If there is available model result information,
It then continues checking for Hydropower Unit state recording whether there is, if it does not exist, then prompting typing set state in log information
It records and exits Current Diagnostic task.If there is Hydropower Unit state recording, then the available of the Hydropower Unit to be diagnosed is recycled
Model result records, and model result record includes model training file and model information, by reading the first row operating parameter letter
Breath obtains model training file content, and tissue generates Hydropower Unit "current" model operation data accordingly, and generates csv to be predicted
Data file, according to model relevant information and file to be predicted, calling model anticipation function carries out fault diagnosis, finally to prediction
As a result it is analyzed and is recorded, i.e., each prediction result compared with the state encoding in set state information, if prediction knot
Fruit is consistent with state encoding, illustrates that the prediction result of the Hydropower Unit is exactly the meaning that the state encoding represents, checks the unit
In status information whether processing is true, if it is true, need result information being recorded in idagnostic logout result, believe
Breath includes Diagnostic Time, Status Name, suggestion takes measures, generating state reason does not record then if NO.Primary pre-
It surveys in result, first is only retained when there is identical prediction result, if prediction result does not appear in set state information
Meeting record log information, and prompt unit status information imperfect, typing asked to be encoded to the set state information of prediction result.
According to one embodiment of present invention, Fig. 3 is that timing failures predict flow chart, i.e., timing failures prediction task is interior
Portion's implementation flow chart.Troubleshooting Flowchart is similar with failure predication flow, and same place is not repeating.Just start check be
Available prediction task, i.e. Prediction program class name whether there is, and then continue to check diagnostic model, model result, set state
Etc. information whether there is.Cycle will predict the model result of Hydropower Unit if all existing, and per diem, week, unite for the period moon three
Average value of corresponding nearest 60 cycle phases of meter with the Hydropower Unit operational parameter data under operating mode.Then time series is used to calculate
Method, 3 week after predicting 60 day, week, the moon using the addition model of long-term trend, seasonal move, cyclical variations, erratic variation
The data of phase.Prediction result is preserved into csv data files to be predicted, according to model relevant information and file to be predicted, is adjusted
Fault diagnosis is carried out with model prediction function, finally prediction result is analyzed and recorded, analysis result is stored in prediction knot
In fruit information, for predicting record queries functional inquiry.
It should be noted that and understand, in the case of the spirit and scope required by not departing from the claims in the present invention, energy
It is enough that various modifications and improvements are made to the present invention of foregoing detailed description.It is therefore desirable to protection technical solution range not by
The limitation of given any specific exemplary teachings.
Claims (8)
1. a kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system, which is characterized in that including：Unit information maintenance module, diagnosis mould
Type maintenance module, diagnostic task maintenance module, idagnostic logout enquiry module, artificial fault diagnosis module；
The unit information maintenance module includes unit information service portion and set state service portion；
The diagnostic model maintenance module includes that diagnostic model is safeguarded, model parameter is arranged, model sample is trained, model result makes
With part；
It is realized to diagnostic model by setting training file, the accuracy rate of analyzing and diagnosing model model sample training part
Selection and calibration；
The method of selection and the calibration to diagnostic model includes：After each data item is selected in model sample training,
Can click model train button, the external trainer function that model training button can be where calling model, and relevant parameter
In incoming function parameter；After external function executes training function, training result, i.e. accuracy rate, sample size, prediction number can be exported
Amount, user can be confirmed whether to preserve the model training result according to the height of accuracy rate；Accuracy rate is clicked 98% or more
Preservation model button preserves model information and model training result；If model accuracy rate is less than 95%, can select
Other training patterns or adjustment model parameter are normalized or adjust model sample training data；
The diagnostic task maintenance module includes that timed task is safeguarded, diagnoses unit maintenance, diagnostic log query portion；
The timed task service portion includes timing failures diagnostic task and timing failures prediction task；The timing failures are examined
Disconnected/prediction task can be obtained according to prefixed time interval timing operation diagnostic model, and by the associated diagnosis unit of diagnostic task
The current operating conditions parameter for taking the Hydropower Unit carries out fault diagnosis and forecast analysis to Hydropower Unit；
The detailed process of the timing failures diagnostic task is：Available diagnostic task is first checked for, i.e. diagnostic program class name is
No presence, diagnostic task information input in timed task maintenance function, and are stored in the diagnostic task table of database, herein
Diagnostic task table records in direct searching database, if it does not exist, then diagnostic task is established in prompt in log information
And Current Diagnostic task is exited, wherein log information is also stored in database table；If there is available diagnostic task information,
It then continues checking for available diagnostic model whether there is, diagnostic model information preservation is safeguarded in database table by diagnostic model
Function typing；If there is no available diagnostic model information, then prompt to establish diagnostic model and exit in log information to work as
Preceding diagnostic task；If there is available diagnostic model information, then continues checking for available model result and whether there is, model knot
Fruit is the model information preserved after model training and training result；If available model result is not present, in log information
Middle prompt preservation model result simultaneously exits Current Diagnostic task；If there is available model result information, then water is continued checking for
Motor group state recording whether there is, if it does not exist, then prompting typing unit state recording in log information and exiting to work as
Preceding diagnostic task；If there is Hydropower Unit state recording, then the available model result record for the Hydropower Unit to be diagnosed is recycled,
Model result record includes model training file and model information, according to model training file content, that is, reads the first row operation
Parameter information organizes Hydropower Unit "current" model operation data, and generates csv data files to be predicted, according to model correlation
Information and file to be predicted, calling model anticipation function carry out fault diagnosis, finally prediction result are analyzed and recorded, i.e.,
Each prediction result compared with the state encoding in set state information, if prediction result is consistent with state encoding, explanation
The prediction result of the Hydropower Unit be exactly the state encoding represent meaning, check in the set state information whether processing is
It is no be it is true, if it is true, need result information being recorded in idagnostic logout result, information includes Diagnostic Time, state name
Claim, suggestion takes measures, generating state reason does not record then if NO；In a prediction result, there is identical prediction
Only retain first when as a result, if prediction result does not appear in the meeting record log information in set state information, and carries
Show that set state information is imperfect, typing is asked to be encoded to the set state information of prediction result；
The timing failures predict that the detailed process of task is：Be first begin to check is available prediction task, i.e., pre- ranging
Sequence class name whether there is, and then continue to check that diagnostic model, model result, set state information whether there is；If all existed
Then cycle will predict the model result of Hydropower Unit, per diem, week, three periods of the moon count nearest 60 cycle phase with the water under operating mode
The average value of motor group operational parameter data；Then time series algorithm is used, including：Long-term trend, seasonal move, cycle become
Dynamic, erratic variation addition model, the data in 3 periods after 60 day of prediction, week, the moon；Prediction result is preserved pre- at waiting for
The csv data files of survey, according to model relevant information and file to be predicted, calling model anticipation function carries out fault diagnosis, most
Prediction result is analyzed and recorded afterwards, analysis result is stored in prediction result information, is looked into for prediction record queries function
It askes；
The idagnostic logout enquiry module includes idagnostic logout query portion and prediction record queries module；
The artificial fault diagnosis module includes artificial fault diagnosis and Artificial Diagnosis query portion.
2. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the unit information is safeguarded
With set state service portion：The typing of Hydropower Unit information or Hydropower Unit state is realized by interface, modification, deletes behaviour
Make；
Wherein：Unit information includes power station title, machine group #, group name, rated power, unit model, commissioning date, system
Make producer, installation unit essential information；Whether set state information includes state encoding, Status Name, handles, suggests taking and arrange
It applies, generating state reason.
3. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the diagnostic model is safeguarded
Part is by interface operation, typing fault diagnosis model information, and fault diagnosis model information includes model name, model class
Whether name training function name, anticipation function name, model description, uses；
Wherein：Model class name is external compiled class name or Service name, and anticipation function name and training function name are outer
Portion's system services compiled custom function name or interface name.
4. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the model result uses
Part selects a highest Hydropower Unit of accuracy rate by the model training result information preserved during the model training of front
The fault diagnosis model used under specific operating mode carries out fault diagnosis and prediction with the diagnostic model.
5. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the diagnosis unit maintenance
Part is by interface operation, and actively selection needs to carry out the Hydropower Unit of fault diagnosis and prediction user, and to being chosen unit
Carry out fault diagnosis and prediction.
6. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the diagnostic log inquiry
Part can preserve system and execute the log information generated when fault diagnosis and failure predication task every time, when user is grasped by interface
When inquiring, to the corresponding log information of user feedback；
The log information content includes task names, logging level, logging time, log content.
7. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the idagnostic logout inquiry
Module includes active inquiry and passive notification portion, wherein active inquiry part is：User is held by interface operation inquiry system
Diagnostic result after row fault diagnosis, diagnostic result content include group name, set state, Diagnostic Time, diagnostic result；Quilt
Dynamic notification portion is：System discovery unit understands automatically prompting user failure happening part, and carry there are when failure during diagnosis
The reason of showing measure that suggestion is taken and may occurring.
8. opening Approach for Hydroelectric Generating Unit Fault Diagnosis system according to claim 1, which is characterized in that the artificial fault diagnosis
Part is active diagnosis of partial：User selects available diagnostic model and uploads diagnostic file, and then system can be to diagnostic file
Carry out fault diagnosis；Wherein, it is desirable that the fault diagnosis file of upload must be with the training file in the fault diagnosis model that uses
Format is consistent, that is, requires have identical columns, the meaning that each column represents is consistent；
The Artificial Diagnosis query portion is to inquire previous artificial fault diagnosis result by interface operation, wherein Artificial Diagnosis
Inquiry content includes the reason of group name, set state, Diagnostic Time, diagnostic result, suggestion take measures, may occur.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510760928.4A CN105278524B (en) | 2015-11-10 | 2015-11-10 | A kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510760928.4A CN105278524B (en) | 2015-11-10 | 2015-11-10 | A kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105278524A CN105278524A (en) | 2016-01-27 |
| CN105278524B true CN105278524B (en) | 2018-10-26 |
Family
ID=55147704
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510760928.4A Active CN105278524B (en) | 2015-11-10 | 2015-11-10 | A kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105278524B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650954A (en) * | 2016-09-18 | 2017-05-10 | 珠海格力电器股份有限公司 | Air conditioner and processing method, device and system of operation data of air conditioner |
| CN107065824B (en) * | 2017-03-07 | 2018-07-17 | 贵州黔源电力股份有限公司 | A kind of Hydropower Unit remote fault diagnosis open platform |
| CN107704933A (en) * | 2017-09-01 | 2018-02-16 | 新疆金风科技股份有限公司 | Wind power generating set fault diagnosis system and method |
| CN109581994A (en) * | 2017-09-28 | 2019-04-05 | 深圳市优必选科技有限公司 | Robot fault diagnosis method and system and terminal equipment |
| CN110309981A (en) * | 2019-07-09 | 2019-10-08 | 华能四川水电有限公司 | A kind of power station Decision-making of Condition-based Maintenance system based on industrial big data |
| CN113295412B (en) * | 2021-05-26 | 2022-10-11 | 华能澜沧江水电股份有限公司 | Method for detecting cause of unbalanced stress of guide bearing of vertical water turbine generator set |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6144923A (en) * | 1997-02-22 | 2000-11-07 | Voith Hydro, Gmbh & Co Kg | Machine diagnosis system |
| KR20090015617A (en) * | 2007-08-09 | 2009-02-12 | 한국전력공사 | System and method for on-line diagnostic of generator-motor |
| CN103366259A (en) * | 2013-07-30 | 2013-10-23 | 国家电网公司 | Method for planning inspection project in hydroelectric production management system |
| CN103366258A (en) * | 2013-07-30 | 2013-10-23 | 国家电网公司 | Method for evaluating indexes of hydroelectric production management system |
| CN103491354A (en) * | 2013-10-10 | 2014-01-01 | 国家电网公司 | System operation monitoring and controlling visual platform |
| CN103728949A (en) * | 2014-01-02 | 2014-04-16 | 贵州电网公司 | Integrated control device of hydroelectric generating set |
| CN104281130A (en) * | 2014-09-22 | 2015-01-14 | 国家电网公司 | Hydroelectric equipment monitoring and fault diagnosis system based on big data technology |
-
2015
- 2015-11-10 CN CN201510760928.4A patent/CN105278524B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6144923A (en) * | 1997-02-22 | 2000-11-07 | Voith Hydro, Gmbh & Co Kg | Machine diagnosis system |
| KR20090015617A (en) * | 2007-08-09 | 2009-02-12 | 한국전력공사 | System and method for on-line diagnostic of generator-motor |
| CN103366259A (en) * | 2013-07-30 | 2013-10-23 | 国家电网公司 | Method for planning inspection project in hydroelectric production management system |
| CN103366258A (en) * | 2013-07-30 | 2013-10-23 | 国家电网公司 | Method for evaluating indexes of hydroelectric production management system |
| CN103491354A (en) * | 2013-10-10 | 2014-01-01 | 国家电网公司 | System operation monitoring and controlling visual platform |
| CN103728949A (en) * | 2014-01-02 | 2014-04-16 | 贵州电网公司 | Integrated control device of hydroelectric generating set |
| CN104281130A (en) * | 2014-09-22 | 2015-01-14 | 国家电网公司 | Hydroelectric equipment monitoring and fault diagnosis system based on big data technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105278524A (en) | 2016-01-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105278524B (en) | A kind of style of opening Approach for Hydroelectric Generating Unit Fault Diagnosis system | |
| Rasmussen | Risk management, adaptation, and design for safety | |
| López-Campos et al. | Modelling using UML and BPMN the integration of open reliability, maintenance and condition monitoring management systems: An application in an electric transformer system | |
| CN112804093B (en) | Centralized scheduling support method and system based on fault capability center | |
| Wang et al. | Intelligent predictive maintenance (IPdM) for elevator service-through CPS, IOT&S and data mining | |
| US20090093902A1 (en) | Method for production scheduling in a manufacturing execution system of a shop floor | |
| Ramesh et al. | Implementation of predictive maintenance systems in remotely located process plants under industry 4.0 scenario | |
| US20240310797A1 (en) | Determining Appropriate Sequences of Actions to Take Upon Operating States of Industrial Plants | |
| Amaya et al. | SIMPREBAL: An expert system for real-time fault diagnosis of hydrogenerators machinery | |
| Borissova et al. | A concept of intelligent e-maintenance decision making system | |
| Guillén López et al. | Prognostics and health management in advanced maintenance systems | |
| Bjorling et al. | Maintenance knowledge management with fusion of CMMS and CM | |
| Oliveira et al. | Combined application of condition-based maintenance and reliability centred maintenance using PFMEA and lean concepts-a case study | |
| Bai et al. | Process Plant Upgradation Using Reliability, Availability, and Maintainability (RAM) Criteria | |
| Laakso et al. | The use of condition monitoring information for maintenance planning and decision-making | |
| Temprado et al. | Knowledge discovery from trouble ticketing reports in a large telecommunication company | |
| Ip et al. | SCADA in an integrated maintenance management system | |
| Alexandru et al. | Building a smart maintenance architecture using smart devices: A web 2.0 based approach | |
| Koitz et al. | Wind Turbine Fault Localization: A Practical Application of Model-Based Diagnosis | |
| Van Blerk | A web-based multilevel framework for condition monitoring of industrial equipment | |
| Seo et al. | Optimal Weapons System Maintenance Using Prognostics and the Interface between an S1000D-based Interactive Electronic Technical Publication and Condition-based Maintenance | |
| CN118348918A (en) | Multi-source data sensing, fusion and predictive analysis method and system for aircraft assembly production line | |
| Berno et al. | A Data Management and Anomaly Detection Solution for the Entertainment Industry | |
| Antoine et al. | Remote services for power generation equipment | |
| Larsson et al. | New solutions for alarm problems |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |